A compressed air cost savings identification model for deep-level mines

Abstract

Eskom has requested a 15% electricity tariff increase over the next three years, which will put an increased burden on the South African mining sector. The cost of electricity directly influences the profitability of the mining industry. Energy efficiency and energy savings initiatives are becoming crucial, as many mines are implementing strategies to combat increasing electricity costs. Compressed air systems are one of the top energy demand systems used by deep-level mines. These systems are also of the most inefficient systems. Therefore deep-level mines are driving awareness to optimise compressed air systems by implementing energy-savings initiatives, which require rigorous planning to launch and implement. Determining the scope for improvement on compressed air systems is also a complex and a time-consuming process which requires multiple personnel, capital and specialised equipment. This study aims to develop a cost-saving identification model which will accurately predict the cost savings potential of a compressed air system on a deep level mine. Research regarding compressed air optimisation, management and demand predictions were conducted to establish a list of methods and tools to be used. No prediction methods were found which could accurately predict the power profile of a compressed air system using minimal input data. Multiple models were constructed to provide a possible solution to the problem using the tools found in research. The models were constructed using statistical data from 29 deep level-mines scattered over South Africa. The data was used to evaluate the relationships between parameters used by previous researchers. The relationships were evaluated and used to construct a prediction model. The cost savings identification model was validated by implementing it on two platinum mines in the Rustenburg area as case studies. Three months' data was used and averaged to evaluate the impact of the implemented initiatives. The estimation model used only the compressed air power baseline as input and produced optimised profile estimations which were 85% and 78% accurate respectively, compared to the actual measurements after optimisation. Due to the dynamic nature of the mining environment, some of the daily estimated results were 100%accurate. These accurate estimations proved that the developed model is capable of aiding in the management of compressed air projects by estimating the savings that can be achieved. The use of the compressed air cost savings identification model decreased the total time to determine feasibility of compressed air energy savings initiatives by 86 weeks, using only two engineering personnel instead of eight. A potential of R 1.05 Million could also have been saved regarding personnel costs if the compressed air energy savings initiatives were not proven feasible. The study aim and study objectives were successfully met while reducing the resources required by a deep-mine to identify the possible cost savings of a compressed air system.